Central Parking System

Libraries…

library(readxl)
library(skimr)

Set path :-

setwd("/Users/himanshusahrawat/MBA projects/CPS/R/")

Read the excel data at sheet zero…

data=read_excel("Central parking dataset.xls", sheet = 1)

New names:

data_s=read_excel("Central parking dataset.xls", sheet = 2)

New names:

Removing unwanted coloumns…

names(data)

 [1] "Vehicle"     "Equipment"   "DateIn"      "Time In"     "DateOut"     "Time Out"    "Amount"     
 [8] "TimeDiff"    "Ticket_Type" "Weekday"     "...11"       "...12"       "...13"      

print("--------------------------------------------------------------------------------")

[1] "--------------------------------------------------------------------------------"

names(data_s)

 [1] "Vehicle"     "Equipment"   "DateIn"      "...4"        "DateOut"     "...6"        "Amount"     
 [8] "TimeDiff"    "Ticket_Type" "Weekday"     "...11"       "...12"       "...13"      

data=data[ , c("Vehicle","Equipment","DateIn","Time In","DateOut","Time Out","Amount","TimeDiff","Ticket_Type","Weekday" )]

data_s=data_s[ , c("Vehicle","Equipment","DateIn","...4","DateOut","...6","Amount","TimeDiff","Ticket_Type","Weekday" )]

Renaming Columns for weekend data…

colnames(data_s)[4] <- "Time In"
colnames(data_s)[6] <- "Time Out"

Summary for weekday data…

skim(data)

── Data Summary ────────────────────────
                           Values
Name                       data  
Number of rows             5000  
Number of columns          10    
_______________________          
Column type frequency:           
  character                2     
  numeric                  4     
  POSIXct                  4     
________________________         
Group variables            None  

observations from weekday data :-

── Data Summary ──────────────────────── Values Name data
Number of rows 5000
Number of columns 10
_______________________
Column type frequency:
character 2
numeric 4
POSIXct 4
________________________
Group variables None

── Variable type: character ────────────────────────────────────────── skim_variable n_missing complete_rate min max empty n_unique 1 Ticket_Type 0 1 6 6 0 2 2 Weekday 0 1 6 9 0 5 whitespace 1 0 2 0

── Variable type: numeric ──────────────────────────────────────────── skim_variable n_missing complete_rate mean sd p0 p25 p50 p75 1 Vehicle 0 1 1 0 1 1 1 1 2 Equipment 0 1 5.63 0.913 5 5 5 6 3 Amount 0 1 36.8 28.5 30 30 30 40 4 TimeDiff 0 1 136. 83.9 2 76 122 179 p100 hist 1 1 ▁▁▇▁▁ 2 8 ▇▅▁▁▂ 3 300 ▇▁▁▁▁ 4 722 ▇▅▁▁▁

── Variable type: POSIXct ──────────────────────────────────────────── skim_variable n_missing complete_rate min
1 DateIn 0 1 2009-07-06 00:00:00 2 Time In 0 1 1899-12-31 00:01:00 3 DateOut 0 1 2009-07-06 00:00:00 4 Time Out 0 1 1899-12-31 00:00:00 max median n_unique 1 2012-04-20 00:00:00 2010-12-28 00:00:00 696 2 1899-12-31 23:58:00 1899-12-31 16:24:00 880 3 2012-04-20 00:00:00 2010-12-28 00:00:00 760 4 1899-12-31 23:59:00 1899-12-31 18:11:00 952

Summary for weekend data…

skim(data_s)

── Data Summary ────────────────────────
                           Values
Name                       data_s
Number of rows             4995  
Number of columns          10    
_______________________          
Column type frequency:           
  character                2     
  numeric                  4     
  POSIXct                  4     
________________________         
Group variables            None  

observation from weekend data ….

── Data Summary ──────────────────────── Values Name data_s Number of rows 4995
Number of columns 10
_______________________
Column type frequency:
character 2
numeric 4
POSIXct 4
________________________
Group variables None

── Variable type: character ────────────────────────────────────────── skim_variable n_missing complete_rate min max empty n_unique 1 Ticket_Type 0 1 4 6 0 3 2 Weekday 0 1 6 8 0 2 whitespace 1 0 2 0

── Variable type: numeric ──────────────────────────────────────────── skim_variable n_missing complete_rate mean sd p0 p25 p50 p75 1 Vehicle 0 1 1 0 1 1 1 1 2 Equipment 0 1 5.95 1.13 5 5 6 6 3 Amount 0 1 62.5 31.4 0 50 50 70 4 TimeDiff 0 1 154. 87.7 1 91 141 200 p100 hist 1 1 ▁▁▇▁▁ 2 9 ▇▆▁▃▁ 3 300 ▇▃▁▁▁ 4 713 ▇▆▁▁▁

── Variable type: POSIXct ──────────────────────────────────────────── skim_variable n_missing complete_rate min
1 DateIn 0 1 2009-07-04 00:00:00 2 Time In 0 1 1899-12-31 00:01:00 3 DateOut 0 1 2009-07-04 00:00:00 4 Time Out 0 1 1899-12-31 00:00:00 max median n_unique 1 2012-03-31 00:00:00 2010-10-30 00:00:00 279 2 1899-12-31 23:58:00 1899-12-31 16:39:00 852 3 2012-04-01 00:00:00 2010-10-30 00:00:00 364 4 1899-12-31 23:59:00 1899-12-31 18:35:00 893

plots for length of stay

weekdays

plot(x = data$TimeDiff, y = data$Amount,
    xlab = "Length of Stay",
    ylab = "Amount",
    main = "Weekday"
)

plot(x = data_s$TimeDiff, y = data_s$Amount,
    xlab = "Length of Stay",
    ylab = "Amount",
    main = "Weekend"
)

library(tidyverse)

Registered S3 methods overwritten by 'dbplyr':
  method         from
  print.tbl_lazy     
  print.tbl_sql      
── Attaching packages ───────────────────────────────────────────────────────────────── tidyverse 1.3.2 ──✔ ggplot2 3.3.6      ✔ purrr   0.3.4 
✔ tibble  3.1.8      ✔ dplyr   1.0.10
✔ tidyr   1.2.1      ✔ stringr 1.4.1 
✔ readr   2.1.3      ✔ forcats 0.5.2 ── Conflicts ──────────────────────────────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag()    masks stats::lag()

#remove the zero and 300 amount
data = data %>% filter(Amount <300)

plot(x = data$TimeDiff, y = data$Amount,
    xlab = "Length of Stay",
    ylab = "Amount",
    main = "Weekday"
)

data_s = data_s %>% filter(Amount <300) %>% filter(Amount>0)

plot(x = data_s$TimeDiff, y = data_s$Amount,
    xlab = "Length of Stay",
    ylab = "Amount",
    main = "Weekend"
)

library(plotly)

Registered S3 method overwritten by 'data.table':
  method           from
  print.data.table     
Registered S3 method overwritten by 'htmlwidgets':
  method           from         
  print.htmlwidget tools:rstudio

Attaching package: ‘plotly’

The following object is masked from ‘package:ggplot2’:

    last_plot

The following object is masked from ‘package:stats’:

    filter

The following object is masked from ‘package:graphics’:

    layout

Plot for count on daily basis.

library(ggplot2)

ggplot(data, aes(x = Weekday)) +
  geom_bar()

ggplot(data_s, aes(x = Weekday)) +
  geom_bar()

Lets check the size of vehicles for both weekdays and weekends….

data %>% filter(Amount>30 & TimeDiff<181)

data_s %>% filter(Amount>60 & TimeDiff<181)

Adding colloumn for timing, if morning (00:00:00 - 12:00:00), afternoon (12:00:00 - 18:00:00), evening (18:00:00 - 00:00:00)

# library(data.table)
 library(readxl)
# install.packages("skimr")
 library("skimr")
# install.packages("tidyverse")
 library("tidyverse")
# install.packages("plotly")
 library("plotly")
# install.packages("lubridate")
 library("lubridate")

Attaching package: ‘lubridate’

The following objects are masked from ‘package:base’:

    date, intersect, setdiff, union

# install.packages("recommenderlab")
 library("recommenderlab")

Loading required package: Matrix

Attaching package: ‘Matrix’

The following objects are masked from ‘package:tidyr’:

    expand, pack, unpack

Loading required package: arules

Attaching package: ‘arules’

The following object is masked from ‘package:dplyr’:

    recode

The following objects are masked from ‘package:base’:

    abbreviate, write

Loading required package: proxy

Attaching package: ‘proxy’

The following object is masked from ‘package:Matrix’:

    as.matrix

The following objects are masked from ‘package:stats’:

    as.dist, dist

The following object is masked from ‘package:base’:

    as.matrix

Loading required package: registry
Registered S3 methods overwritten by 'registry':
  method               from 
  print.registry_field proxy
  print.registry_entry proxy

class(data$`Time In`)

[1] "POSIXct" "POSIXt" 

#hour(data$`Time In`)

# Multiple conditions when adding new column to dataframe:
data=data %>% mutate(Timing =
                     case_when((hour(`Time In`) >=00) & (hour(`Time In`)) <12 ~ "Morning", 
                               (hour(`Time In`) >=12) & (hour(`Time In`)) <18 ~ "Afternoon",
                               (hour(`Time In`) >=18) & (hour(`Time In`)) <24 ~ "Evening"))

data_s=data_s %>% mutate(Timing =
                     case_when((hour(`Time In`) >=00) & (hour(`Time In`)) <12 ~ "Morning", 
                               (hour(`Time In`) >=12) & (hour(`Time In`)) <18 ~ "Afternoon",
                               (hour(`Time In`) >=18) & (hour(`Time In`)) <24 ~ "Evening"))

tw = data %>% 
  group_by(Timing) %>%
  summarise(count =n())
  

tw

fig <- plot_ly(tw, labels = ~Timing, values = ~count, type = 'pie')

fig <- fig %>% layout(title = 'Weekday Timing',
         xaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE),
         yaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE))
fig

people are coming in afternoon

tw = data_s %>% 
  group_by(Timing) %>%
  summarise(count =n())


fig <- plot_ly(tw, labels = ~Timing, values = ~count, type = 'pie')

fig <- fig %>% layout(title = 'Weekend Timing',
         xaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE),
         yaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE))
fig

During Saturday Sunday , most of the person are coming in afternoon.

Distribution of time spent by vehicles —–

library(ggplot2)

## Basic histogram from the vector "rating". Each bin is .5 wide.
## These both result in the same output:
ggplot(data, aes(x=TimeDiff)) + geom_histogram(binwidth=10)

# qplot(dat$rating, binwidth=.5)

# Draw with black outline, white fill
ggplot(data, aes(x=TimeDiff)) +
    geom_histogram(binwidth=10, colour="black", fill="white")

# Density curve
ggplot(data, aes(x=TimeDiff)) + geom_density()

# Histogram overlaid with kernel density curve
ggplot(data, aes(x=TimeDiff)) + 
    geom_histogram(aes(y=..density..),      # Histogram with density instead of count on y-axis
                   binwidth=10,
                   colour="black", fill="white") +
    geom_density(alpha=.2, fill="#FF6666")  # Overlay with transparent density plot

Checking Skew Value

install.packages("moments")

trying URL 'https://cran.rstudio.com/bin/macosx/contrib/4.2/moments_0.14.1.tgz'
Content type 'application/x-gzip' length 54374 bytes (53 KB)
==================================================
downloaded 53 KB

The downloaded binary packages are in
    /var/folders/hk/_k0j74ld46j579pyq03bync00000gn/T//RtmpNqXUjs/downloaded_packages

library(moments)

skewness(data$TimeDiff)

[1] 1.40518

.If the skewness is between -0.5 and 0.5, the data are fairly symmetrical . If the skewness is between -1 and — 0.5 or between 0.5 and 1, the data are moderately skewed . If the skewness is less than -1 or greater than 1, the data are highly skewed

DISTRIBUTION FOR WEEKEND

## Basic histogram from the vector "rating". Each bin is .5 wide.
## These both result in the same output:
ggplot(data_s, aes(x=TimeDiff)) + geom_histogram(binwidth=10)

# qplot(dat$rating, binwidth=.5)

# Draw with black outline, white fill
ggplot(data_s, aes(x=TimeDiff)) +
    geom_histogram(binwidth=10, colour="black", fill="white")

# Density curve
ggplot(data_s, aes(x=TimeDiff)) + geom_density()

# Histogram overlaid with kernel density curve
ggplot(data_s, aes(x=TimeDiff)) + 
    geom_histogram(aes(y=..density..),      # Histogram with density instead of count on y-axis
                   binwidth=10,
                   colour="black", fill="white") +
    geom_density(alpha=.2, fill="#FF6666")  # Overlay with transparent density plot

skewness(data_s$TimeDiff)

[1] 1.056578

Q.4

People tends to stay longer during weekends.

print("Average stay During weekday")

[1] "Average stay During weekday"

print(mean(data$TimeDiff))

[1] 135.4679

print("Average stay During weekend")

[1] "Average stay During weekend"

print(mean(data_s$TimeDiff))

[1] 151.9696

Yes people tends to stay longer during weekends.

Q.5

data_s=data_s %>% mutate(Timing =
                     case_when((hour(`Time In`) >=00) & (hour(`Time In`)) <14 ~ "Morning", 
                               (hour(`Time In`) >=14) & (hour(`Time In`)) <18 ~ "Afternoon",
                               (hour(`Time In`) >=18) & (hour(`Time In`)) <24 ~ "Evening"))

tw = data_s %>% 
  group_by(Timing) %>%
  summarise(count =n())


fig <- plot_ly(tw, labels = ~Timing, values = ~count, type = 'pie')

fig <- fig %>% layout(title = 'Weekend Timing',
         xaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE),
         yaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE))
fig

Q6.. If charges for first 2 hr is 20 and 10 for every additional hour… financial effect.

new_price <- function(a) {
  b=as.integer(a[8])-120
  amount=20
  while (b>0) {
    amount=amount+10
    b=b-60
  }
  return(amount)
  }

data$newprice = apply(data,1,new_price)

sum(data$Amount)-sum(data$newprice)

[1] 24440

loss of 24440

---

q.8 analyst claim- aternate hypo- average ocupancy is not utmost 2 hr for weekdays. alpha - 0.05 . pop var=sam var

data_sample=sample_n(data, 500)

## Basic histogram from the vector "rating". Each bin is .5 wide.
## These both result in the same output:
ggplot(data_sample, aes(x=TimeDiff)) + geom_histogram(binwidth=.01)

# qplot(dat$rating, binwidth=.5)

# Draw with black outline, white fill
ggplot(data_sample, aes(x=TimeDiff)) +
    geom_histogram(binwidth=.01, colour="black", fill="white")

# Density curve
ggplot(data_sample, aes(x=TimeDiff)) + geom_density()

# Histogram overlaid with kernel density curve
ggplot(data_sample, aes(x=TimeDiff)) + 
    geom_histogram(aes(y=..density..),      # Histogram with density instead of count on y-axis
                   binwidth=.01,
                   colour="black", fill="white") +
    geom_density(alpha=.2, fill="#FF6666")

install.packages("BSDA")

library(BSDA)

h null : mean <=120 (claimed value) h alter : mean >120 (opposite)

The alternative hypothesis in each case indicates the direction of divergence of the population mean for x (or difference of means for x and y) from mu (i.e., “greater”, “less”, “two.sided”).

data_sample=sample_n(data, 500)
z.test(x=data_sample$TimeDiff, mu=120, sigma.x=var(data$TimeDiff),alternative = "greater")

The test statistic for the one sample z-test is 0.054417 and the corresponding p-value is 0.4783.

Since this p-value is not less than .05, we have have sufficient evidence to reject the null hypothesis.

Thank you !!!!!